In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.
The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.
In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!
We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.
Make sure that you've downloaded the required human and dog datasets:
Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dogImages.
Download the human dataset. Unzip the folder and place it in the home directory, at location /lfw.
Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.
In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.
import numpy as np
from glob import glob
# load filenames for human and dog images
human_files = np.array(glob("lfw/*/*"))
dog_files = np.array(glob("dogImages/*/*/*"))
# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
There are 13233 total human images. There are 8351 total dog images.
In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.
OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.
import cv2
import matplotlib.pyplot as plt
%matplotlib inline
# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')
# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# find faces in image
faces = face_cascade.detectMultiScale(gray)
# print number of faces detected in the image
print('Number of faces detected:', len(faces))
# get bounding box for each detected face
for (x,y,w,h) in faces:
# add bounding box to color image
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Number of faces detected: 2
Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.
In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.
We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
img = cv2.imread(img_path)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray)
return len(faces) > 0
Question 1: Use the code cell below to test the performance of the face_detector function.
human_files have a detected human face? dog_files have a detected human face? Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.
Answer: The answer is given in the console output below.
from tqdm import tqdm, trange
import time
start_time = time.time()
human_files_short = human_files[:100]
dog_files_short = dog_files[:100]
#-#-# Do NOT modify the code above this line. #-#-#
human_tp = 0
dog_fp = 0
for i in trange(100):
human_tp += int(face_detector(human_files_short[i]) == True)
dog_fp += int(face_detector(dog_files_short[i]) == True)
print("What percentage of the first 100 images in human_files have a detected human face? {:.0f}%"
.format(human_tp))
print("What percentage of the first 100 images in dog_files have a detected human face? {:.0f}%"
.format(dog_fp))
print("Runtime: %.2f seconds" % (time.time() - start_time))
100%|██████████| 100/100 [00:08<00:00, 12.05it/s]
What percentage of the first 100 images in human_files have a detected human face? 99% What percentage of the first 100 images in dog_files have a detected human face? 13% Runtime: 8.30 seconds
We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
pass
In this section, we use a pre-trained model to detect dogs in images.
The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.
import torch
import torchvision.models as models
# define VGG16 model
VGG16 = models.vgg16(pretrained=True)
# check if CUDA is available
use_cuda = torch.cuda.is_available()
# move model to GPU if CUDA is available
if use_cuda:
VGG16 = VGG16.cuda()
Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.
In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.
Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.
from PIL import Image
import torchvision.transforms as transforms
# Set PIL to be tolerant of image files that are truncated.
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
def VGG16_predict(img_path):
'''
Use pre-trained VGG-16 model to obtain index corresponding to
predicted ImageNet class for image at specified path
Args:
img_path: path to an image
Returns:
Index corresponding to VGG-16 model's prediction
'''
image = Image.open(img_path)
transform = transforms.Compose([
transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
img = transform(image)
img = img.unsqueeze(0)
img = img.cuda()
out = VGG16.forward(img)
_, pred = torch.max(out, 1)
return pred.item()
# test functionality
img = dog_files[4156]
index = VGG16_predict(img)
display(Image.open(img))
print(img, index)
dogImages/train/023.Bernese_mountain_dog/Bernese_mountain_dog_01651.jpg 239
While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).
Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
index = VGG16_predict(img_path)
return True if 151 <= index <= 268 else False
# test functionality
dog = dog_files[0]
human = human_files[0]
print(dog_detector(dog))
print(dog_detector(human))
True False
Question 2: Use the code cell below to test the performance of your dog_detector function.
human_files_short have a detected dog? dog_files_short have a detected dog?Answer: The answer is given in the console output below.
start_time = time.time()
human_tp = 0
dog_fp = 0
for i in trange(100):
human_tp += int(dog_detector(human_files_short[i]) == True)
dog_fp += int(dog_detector(dog_files_short[i]) == True)
print("What percentage of the images in human_files_short have a detected dog? {:.0f}%"
.format(human_tp))
print("What percentage of the images in dog_files_short have a detected dog? {:.0f}%"
.format(dog_fp))
print("Runtime: %.2f seconds" % (time.time() - start_time))
100%|██████████| 100/100 [00:02<00:00, 40.95it/s]
What percentage of the images in human_files_short have a detected dog? 2% What percentage of the images in dog_files_short have a detected dog? 100% Runtime: 2.44 seconds
We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.
pass
Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.
We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.
| Brittany | Welsh Springer Spaniel |
|---|---|
![]() |
![]() |
It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).
| Curly-Coated Retriever | American Water Spaniel |
|---|---|
![]() |
![]() |
Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.
| Yellow Labrador | Chocolate Labrador | Black Labrador |
|---|---|---|
![]() |
![]() |
![]() |
We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.
Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!
import os
import torchvision.transforms as transforms
from torchvision import datasets
num_workers = 0
batch_size = 20
transform_train = transforms.Compose([transforms.RandomRotation(5),
transforms.Resize((256, 256)),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5]),
])
transform = transforms.Compose([transforms.Resize((256, 256)),
transforms.ToTensor(),
transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5]),
])
train_data = datasets.ImageFolder(root='dogImages/train', transform=transform_train)
valid_data = datasets.ImageFolder(root='dogImages/valid', transform=transform)
test_data = datasets.ImageFolder(root='dogImages/test', transform=transform)
train_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
valid_loader = torch.utils.data.DataLoader(valid_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
test_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
loaders_scratch = {}
loaders_scratch["train"] = train_loader
loaders_scratch["valid"] = valid_loader
loaders_scratch["test"] = test_loader
Question 3: Describe your chosen procedure for preprocessing the data.
Answer: The code randomly rotates, resizes and flips the input images for the training dataset. Randomizing training data helps to prevent overfitting. Also, I am normalizing the pixel values. Normalization is used to keep the network weights near zero which makes the backpropagation process more stable during training. For the validation and test datasets, I choose to only resize and normalize the input images. I use an input size of 256-by-256 pixels for all datasets, because the provided input images vary in size. Keeping a static input size will help me to validate the classification performance throughout the complete dataset. The reason why I am not introducing any randomness during the validation and test steps is that we want to check how well the neural network is able to generalize on unseen data.
Remarks: It turned out that applying randomness during the pre-processing step did not have a significant positive effect on the image classification performance. I conclude that it is more important to design a suitable model, rather than applying data augmentation on input images for training alone.
import matplotlib.pyplot as plt
%matplotlib inline
def plot_dataset(loader):
# obtain one batch of training images
data_iter = iter(loader)
images, labels = data_iter.next()
# convert images to numpy for display
images = images.numpy()
# plot the images in the batch, along with the corresponding labels
fig = plt.figure(figsize=(25, 4))
# display 20 images
for idx in np.arange(20):
ax = fig.add_subplot(2, int(20/2), idx+1, xticks=[], yticks=[])
img = images[idx]
img = img / 2 + 0.5 # un-normalize
img = np.clip(img, 0, 1) # clip inputs to interval edges
plt.imshow(np.transpose(img, (1, 2, 0)))
ax.set_title("Class " + str(labels[idx].item()))
plot_dataset(loaders_scratch["train"])
Create a CNN to classify dog breed. Use the template in the code cell below.
import torch.nn as nn
import torch.nn.functional as F
# define the CNN architecture
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
self.conv1 = nn.Conv2d(3, 16, 3, padding=1)
self.conv2 = nn.Conv2d(16, 32, 3, padding=1)
self.conv3 = nn.Conv2d(32, 64, 3, padding=1)
self.conv4 = nn.Conv2d(64, 128, 3, padding=1)
self.conv5 = nn.Conv2d(128, 256, 3, padding=1)
self.pool = nn.MaxPool2d(2, 2)
self.fc1 = nn.Linear(256 * 8 * 8, 512)
self.fc2 = nn.Linear(512, 133) # make a prediction between 133 classes of dog breeds
self.dropout = nn.Dropout(p=0.25)
def forward(self, x):
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = self.pool(F.relu(self.conv3(x)))
x = self.pool(F.relu(self.conv4(x)))
x = self.pool(F.relu(self.conv5(x)))
x = x.view(-1, 256 * 8 * 8) # input image size is 8x8 with a depth of 265
x = self.dropout(x)
x = F.relu(self.fc1(x))
x = self.dropout(x)
x = self.fc2(x)
return x
#-#-# You do NOT have to modify the code below this line. #-#-#
# instantiate the CNN
model_scratch = Net()
# move tensors to GPU if CUDA is available
if use_cuda:
model_scratch.cuda()
# test forward pass
data_iter = iter(loaders_scratch["train"])
images, labels = data_iter.next()
images = images.cuda()
model_scratch(images)
tensor([[ 0.0178, 0.0143, -0.0374, ..., 0.0398, -0.0219, -0.0306],
[ 0.0189, 0.0210, -0.0423, ..., 0.0303, -0.0144, -0.0350],
[ 0.0130, 0.0186, -0.0416, ..., 0.0382, -0.0252, -0.0320],
...,
[ 0.0140, 0.0211, -0.0493, ..., 0.0307, -0.0193, -0.0349],
[ 0.0148, 0.0123, -0.0454, ..., 0.0378, -0.0159, -0.0410],
[ 0.0113, 0.0149, -0.0394, ..., 0.0307, -0.0282, -0.0297]],
device='cuda:0', grad_fn=<AddmmBackward>)
Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.
Answer: I choose a CNN architecture which is very similar to what I have been introduced to during the Udacity online course. This way, I am better able to compare the classification performances among all neural networks I have self-constructed so far. I have chosen five convolutional layers in total, because after analyzing the image datasets, I feel like the complexity of the images is quite high. The images strongly vary in size, the dogs themselves are depicted in different sizes, some dogs have their head tilted, some dog breeds look very identical or have different color of fur, etc. But I hope that the more hidden layers I include in the network for the feature detection, the more complex patterns this network will be able to detect. The classifier in the fully-connected network part will receive a 20x256x8x8 input tensor. In general, my concern is overfitting based on the explanation concerning the dataset complexity. If my model fails to achieve the 10% accuracy requirement, then I will take a closer look at how the VGG16 architecture is designed and derive appropriate corrections. Maybe increasing the dropout probability from 0.25 to 0.5 helps to decrease the overfitting effect.
Remarks: It turned out that increasing the number of hidden output nodes of self.fc1 from 256 to 512 had a positive effect on the final test accuracy. It improved from 11% to 15%.
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.
import torch.optim as optim
### select loss function
criterion_scratch = nn.CrossEntropyLoss()
### select optimizer
optimizer_scratch = optim.Adam(model_scratch.parameters(), lr=0.001)
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.
# the following import is required for training to be robust to truncated images
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
"""returns trained model"""
# initialize start time
start_time = time.time()
# initialize tracker for minimum validation loss
valid_loss_min = np.Inf
for epoch in range(1, n_epochs+1):
# initialize variables to monitor training and validation loss
train_loss = 0.0
valid_loss = 0.0
###################
# train the model #
###################
model.train()
for batch_idx, (data, target) in enumerate(loaders["train"]):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# find the loss and update the model parameters accordingly
optimizer.zero_grad()
output = model(data)
loss = criterion(output, target)
loss.backward()
optimizer.step()
train_loss += loss.item() * data.size(0)
######################
# validate the model #
######################
with torch.no_grad():
model.eval()
for batch_idx, (data, target) in enumerate(loaders["valid"]):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# update the average validation loss
output = model(data)
loss = criterion(output, target)
valid_loss += loss.item() * data.size(0)
# calculate average losses
train_loss = train_loss/len(loaders["train"].dataset)
valid_loss = valid_loss/len(loaders["valid"].dataset)
# print training/validation statistics every five epochs
if epoch % 5 == 0:
print("Epoch: {} \tTraining Loss: {:.4f} \tValidation Loss: {:.4f}".format(epoch, train_loss, valid_loss))
print("Runtime: %.2f minutes" % ((time.time() - start_time) / 60.0))
# save model
if valid_loss <= valid_loss_min:
print("Validation loss decreased from {:.4f} to {:.4f}, saving model.".format(valid_loss_min, valid_loss))
torch.save(model.state_dict(), save_path)
valid_loss_min = valid_loss
# print total runtime
print("Total Runtime: %.2f minutes" % ((time.time() - start_time) / 60.0))
# return trained model
return model
# train the model
model_scratch = train(30, loaders_scratch, model_scratch, optimizer_scratch, criterion_scratch, use_cuda, 'model_scratch.pt')
Validation loss decreased from inf to 4.8716, saving model. Validation loss decreased from 4.8716 to 4.7011, saving model. Validation loss decreased from 4.7011 to 4.5634, saving model. Validation loss decreased from 4.5634 to 4.4783, saving model. Epoch: 5 Training Loss: 4.3938 Validation Loss: 4.3060 Runtime: 7.20 minutes Validation loss decreased from 4.4783 to 4.3060, saving model. Validation loss decreased from 4.3060 to 4.1352, saving model. Validation loss decreased from 4.1352 to 4.0130, saving model. Validation loss decreased from 4.0130 to 3.9227, saving model. Validation loss decreased from 3.9227 to 3.8853, saving model. Epoch: 10 Training Loss: 3.4542 Validation Loss: 3.7778 Runtime: 14.46 minutes Validation loss decreased from 3.8853 to 3.7778, saving model. Validation loss decreased from 3.7778 to 3.7372, saving model. Epoch: 15 Training Loss: 2.5735 Validation Loss: 3.8112 Runtime: 21.84 minutes Epoch: 20 Training Loss: 1.8201 Validation Loss: 4.1490 Runtime: 29.15 minutes Epoch: 25 Training Loss: 1.2915 Validation Loss: 4.5415 Runtime: 36.40 minutes Epoch: 30 Training Loss: 1.0037 Validation Loss: 5.0206 Runtime: 43.64 minutes Total Runtime: 43.64 minutes
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
<All keys matched successfully>
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.
def test(loaders, model, criterion, use_cuda):
# monitor test loss and accuracy
test_loss = 0.
correct = 0.
total = 0.
model.eval()
for batch_idx, (data, target) in enumerate(loaders['test']):
# move to GPU
if use_cuda:
data, target = data.cuda(), target.cuda()
# forward pass: compute predicted outputs by passing inputs to the model
output = model(data)
# calculate the loss
loss = criterion(output, target)
# update average test loss
test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
# convert output probabilities to predicted class
pred = output.data.max(1, keepdim=True)[1]
# compare predictions to true label
correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
total += data.size(0)
print('Test Loss: {:.6f}'.format(test_loss))
print('Test Accuracy: %2d%% (%2d/%2d)' % (100. * correct / total, correct, total))
# test the model
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Test Loss: 3.820366 Test Accuracy: 15% (126/836)
You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.
Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).
If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.
num_workers = 0
batch_size = 20
transform_train = transforms.Compose([transforms.Resize((256, 256)),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])
transform = transforms.Compose([transforms.Resize((256, 256)),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])
train_data = datasets.ImageFolder(root='dogImages/train', transform=transform_train)
valid_data = datasets.ImageFolder(root='dogImages/valid', transform=transform)
test_data = datasets.ImageFolder(root='dogImages/test', transform=transform)
train_loader = torch.utils.data.DataLoader(train_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
valid_loader = torch.utils.data.DataLoader(valid_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
test_loader = torch.utils.data.DataLoader(test_data, batch_size=batch_size, num_workers=num_workers, shuffle=True)
loaders_transfer = {}
loaders_transfer["train"] = train_loader
loaders_transfer["valid"] = valid_loader
loaders_transfer["test"] = test_loader
for loader in loaders_transfer:
print(loaders_transfer[loader].dataset)
Dataset ImageFolder
Number of datapoints: 6680
Root Location: dogImages/train
Transforms (if any): Compose(
Resize(size=(256, 256), interpolation=PIL.Image.BILINEAR)
CenterCrop(size=(224, 224))
ToTensor()
Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
)
Target Transforms (if any): None
Dataset ImageFolder
Number of datapoints: 835
Root Location: dogImages/valid
Transforms (if any): Compose(
Resize(size=(256, 256), interpolation=PIL.Image.BILINEAR)
CenterCrop(size=(224, 224))
ToTensor()
Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
)
Target Transforms (if any): None
Dataset ImageFolder
Number of datapoints: 836
Root Location: dogImages/test
Transforms (if any): Compose(
Resize(size=(256, 256), interpolation=PIL.Image.BILINEAR)
CenterCrop(size=(224, 224))
ToTensor()
Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
)
Target Transforms (if any): None
plot_dataset(loaders_transfer["train"])
Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.
import torchvision.models as models
import torch.nn as nn
model_transfer = models.resnet101(pretrained=True)
for param in model_transfer.parameters():
param.requires_grad = False
print(model_transfer.fc)
model_transfer.fc = nn.Linear(model_transfer.fc.in_features, 133) # make a prediction between 133 classes of dog breeds
print(model_transfer.fc)
for param in model_transfer.fc.parameters():
print(param.requires_grad) # check if the last couple of parameters are not frozen
if use_cuda:
model_transfer = model_transfer.cuda()
Linear(in_features=2048, out_features=1000, bias=True) Linear(in_features=2048, out_features=133, bias=True) True True
Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.
Answer: I use a model which is known to perform well on the given image classification task. ResNet has been trained on the ImageNet database which includes images of different dog breeds. By applying transfer learning on a pretrained ResNet model, I achieve state-of-the-art classification performance results for the given image dataset. ResNet consists of a feature extraction and a classifier part. By freezing all parameters of the feature extraction part, and by optimizing the weights of the last linear layer only, I will create a CNN tuned to classify dog breeds. I use ResNet101, because I feel like with a 101-layer architecture, I am guaranteed to exceed the 60% accuracy requirement mentioned below. The number of layers do not matter in terms of training runtime, because as said before, the only part which is optimized is the last linear layer of the CNN.
Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.Adam(model_transfer.fc.parameters(), lr=0.001)
Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.
# train the model
model_transfer = train(30, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')
Validation loss decreased from inf to 0.7099, saving model. Validation loss decreased from 0.7099 to 0.6104, saving model. Validation loss decreased from 0.6104 to 0.5476, saving model. Epoch: 5 Training Loss: 0.2982 Validation Loss: 0.5334 Runtime: 7.88 minutes Validation loss decreased from 0.5476 to 0.5334, saving model. Epoch: 10 Training Loss: 0.1952 Validation Loss: 0.6146 Runtime: 15.78 minutes Epoch: 15 Training Loss: 0.1299 Validation Loss: 0.6018 Runtime: 23.66 minutes Epoch: 20 Training Loss: 0.1191 Validation Loss: 0.7293 Runtime: 31.54 minutes Epoch: 25 Training Loss: 0.0879 Validation Loss: 0.6738 Runtime: 39.44 minutes Epoch: 30 Training Loss: 0.1009 Validation Loss: 0.7618 Runtime: 47.34 minutes Total Runtime: 47.34 minutes
# load the model that got the best validation accuracy
model_transfer.load_state_dict(torch.load('model_transfer.pt'))
<All keys matched successfully>
Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.
# test the model
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
Test Loss: 0.540616 Test Accuracy: 85% (714/836)
Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.
# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in loaders_transfer['train'].dataset.classes]
def predict_breed_transfer(img_path):
# load the image and return the predicted breed
img = Image.open(img_path)
img = transforms.ToTensor()(img).unsqueeze_(0)
img = img.cuda()
output = model_transfer(img)
pred = output.data.max(1, keepdim=True)[1]
return class_names[pred.item()]
dog1 = predict_breed_transfer("dogImages/test/076.Golden_retriever/Golden_retriever_05221.jpg")
dog2 = predict_breed_transfer("dogImages/test/027.Bloodhound/Bloodhound_01885.jpg")
print(dog1)
print(dog2)
Golden retriever Bloodhound
Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,
You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and dog_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.
Some sample output for our algorithm is provided below, but feel free to design your own user experience!

# handle cases for a human face, dog, and neither
def run_app(img_path):
dog = dog_detector(img_path)
human = face_detector(img_path)
img = Image.open(img_path)
if dog:
display(img)
print("Dog! And the breed is {0}.\n\n".format(predict_breed_transfer(img_path)))
elif human:
display(img)
print("Human! And looking like a {}.\n\n".format(predict_breed_transfer(img_path)))
else:
display(img)
print("Neither human nor dog is detected. Unknown error!\n\n")
run_app("lfw/Aaron_Pena/Aaron_Pena_0001.jpg")
run_app("dogImages/test/027.Bloodhound/Bloodhound_01885.jpg")
run_app("images/external/basketball.jpg")
Human! And looking like a Poodle.
Dog! And the breed is Bloodhound.
Neither human nor dog is detected. Unknown error!
In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?
Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.
Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.
Answer: In general, I am quite happy with the results. 1) By extending the image dataset for training, I could achieve better dog breed classification results. The CNN yields only a test accuracy of 85% after transfer learning. This is a concern for similar looking dog breeds as shown below, because the second dog breed is classified incorrectly: Norwegian Buhund instead of German Shepherd. 2) In order to avoid overfitting, Dropout layers in the final classifier part could be included. Further data augmentation techniques could be used for pre-processing, so that the CNN is able to generalize better. 3) It does not make sense that humans are classified in terms of a dog breed. The current dog detector is enforcing a dog breed on any input image. So, I would add another class called Human with corresponding training images. Hopefully, the CNN would learn during training how to distinguish between dogs and humans in the end.
import random
# following images were taken from https://unsplash.com
dogs = ["images/external/dog1.jpg", "images/external/dog2.jpg"]
humans = ["images/external/human1.jpg", "images/external/human2.jpg"]
others = ["images/external/other1.jpg", "images/external/other2.jpg", "images/external/other3.jpg"]
for dog in dogs:
run_app(dog)
for human in humans:
run_app(human)
for other in others:
run_app(other)
Dog! And the breed is Golden retriever.
Dog! And the breed is Norwegian buhund.
Human! And looking like a Lhasa apso.
Human! And looking like a German pinscher.
Neither human nor dog is detected. Unknown error!
Neither human nor dog is detected. Unknown error!
Human! And looking like a Poodle.